Skip to main content

CA AB 2013

Generative Artificial Intelligence: Training Data Transparency

Requires GenAI developers to publish documentation about training datasets including sources, data types, copyright status, personal information inclusion, and processing methods.

Jurisdiction

California

Enacted

Sep 28, 2024

Effective

Jan 1, 2026

Enforcement

Not specified (likely CA AG under unfair competition law)

Signed September 28, 2024; effective January 1, 2026

CA Legislature

Why It Matters

Addresses training data transparency concerns. Relevant for IP/copyright discussions. Relatively light requirements (documentation only).

Recent Developments

Signed September 2024. Focuses on transparency about training data - relevant to copyright and data governance debates.

At a Glance

Applies to

Foundation Model

Harms addressed

Requires

Who Must Comply

  • Developers of GenAI systems publicly available to Californians
  • Developers substantially modifying GenAI systems

Safety Provisions

  • High-level summary of training datasets
  • Sources or owners of datasets
  • Alignment with intended purpose
  • Number and types of data points
  • Copyright/trademark/patent/public domain status
  • Whether datasets purchased or licensed
  • Personal information or aggregate consumer data inclusion
  • Synthetic data usage
  • Data collection timeframes
  • Cleaning/processing methods

Exemptions

National Security/Military

National security/military/defense systems (federal entities only)

  • • Federal entity
  • • National security purpose

Aircraft Operation

Systems for aircraft operation

  • • Aircraft operation purpose

Security/Integrity Systems

Systems solely for security or integrity purposes

  • • Security/integrity purpose only

Compliance & Enforcement

Key Dates

Jan 1, 2026

Training data documentation must be posted on website

Penalties

Not specified in statute

View on map

California

Focus Areas

Algorithmic accountability

Cite This

APA

California. (2024). Generative Artificial Intelligence: Training Data Transparency.

Related Regulations

In Effect US-CA

CA SB 53

First US frontier AI transparency law. Requires large AI developers (>$500M revenue) to publish governance frameworks, submit quarterly risk reports, and report critical safety incidents. Applies to models trained with >10^26 FLOP.

In Effect US-CA

CA CPPA ADMT

California Privacy Protection Agency regulations establishing consumer rights and business obligations for Automated Decision-Making Technology (ADMT) that makes significant decisions including healthcare. Requires pre-use notice, opt-out rights, access rights, appeal rights, and risk assessments.

Enacted US-NY

NY RAISE Act

Requires large AI developers of frontier models operating in New York to create safety protocols, report critical incidents within 72 hours, conduct annual reviews, and undergo independent audits. Creates dedicated DFS office funded by developer fees.

Enacted US-TX

TX Healthcare AI Law

Requires healthcare practitioners using AI for diagnosis to review all AI-generated records and disclose AI use to patients. Mandates EHR data localization (Texas patient data must be physically stored in US). Applies to covered entities and third-party vendors.

Pending US-LA

LA Healthcare AI Act

Regulates use of artificial intelligence by healthcare providers in Louisiana. Permits AI for administrative tasks but prohibits AI from making treatment/diagnosis decisions without licensed professional review, directly interacting with patients on treatment matters, or generating therapeutic recommendations without professional approval.

Enacted US-VT

VT AADC

Vermont design code structured to be more litigation-resistant: focuses on data processing harms rather than content-based restrictions. AG rulemaking authority begins July 2025.

Last updated January 23, 2026. Verify against primary sources before relying on this information.